tensions in the epistemic domain and claims of no-knowledge: a … · 2017-03-13 · tensions in...
TRANSCRIPT
Available online at www.sciencedirect.com
ScienceDirect
Tensions in the epistemic domain and claims ofno-knowledge: A study of Swedish medical interaction
www.elsevier.com/locate/pragmaJournal of Pragmatics 106 (2016) 129--147
Jan Lindström a,*, Susanna Karlsson b
aDepartment of Finnish, Finno-Ugrian and Scandinavian Studies, P.O. Box 24, 00014 University of Helsinki, Finlandb Language Council of Sweden, Institute for Language and Folklore, P.O. Box 20057, SE-10460 Stockholm, Sweden
Available online 19 July 2016
Abstract
This article analyzes the halts in sequential progressivity that are caused by claims of no-knowledge in Swedish medical interaction.The focus is on responsive turns and turn-constructional units that are prefaced by the epistemic disclaimer jag vet inte ‘I don’t know’. Weargue that this use of epistemic disclaimers does not primarily display the speaker’s lack of knowledge, but that their presence signalsinteractional problems that are contingent on epistemic asymmetries between the participants. Patient replies that contain an epistemicdisclaimer are nonconforming responses and they therefore resist something about the question: the presupposed access to knowledgeor the rights to knowledge. The present analysis demonstrates that epistemic tensions, especially in lay--professional interaction, arehandled by the lay party using epistemic disclaimers. These can initiate a shift in epistemic posture toward a more independent, morepersonally accurate formulation of knowledge that somehow contrasts with the professional party’s assumptions.© 2016 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creative-commons.org/licenses/by-nc-nd/4.0/).
Keywords: Grammar-in-interaction; Epistemic disclaimers; Epistemic stance; Epistemic asymmetry; Medical interaction; (Finland) Swedish
1. Introduction
Sequential progressivity is one of the key features of smooth social interaction. The components of the turn-takingsystem ‘‘provide for the possibility of transitions with no gap’’ (Sacks et al., 1974:708). At their completion, turns at talk alsoserve to evoke a relevant next action by the other party -- a principle that is most obvious in the organization of adjacencypairs (Sacks, 1987). For example, a request for information makes an answer that provides a piece of appropriateinformation relevant as the next action (Stivers, 2011:104). Halts in progressivity nonetheless occur, such as when therecipient of a request has trouble hearing or understanding the first action and produces a What? instead of the relevantanswer. These types of responses then initiate repair sequences, which mark a break in the ongoing course of action, butwhich eventually work to mend the break in intersubjectivity and restore progressivity (Heritage, 2007; Sacks et al., 1977).
When a speaker uses responses to claim a lack of knowledge, this is arguably an interactional move that mayeffectively hinder the progressivity. No-knowledge responses to questions share one quality of non-answers in that theyare contributions that fail to deliver the information that the questioner is seeking (Stivers and Robinson, 2006). AsHutchby (2002) demonstrates, the non-cooperative feature of no-knowledge responses may be used strategically toavoid talking about specific topics (see also Tsui, 1991; Weatherall, 2011). The objective of this study is to investigate
* Corresponding author.E-mail addresses: [email protected] (J. Lindström), [email protected] (S. Karlsson).
http://dx.doi.org/10.1016/j.pragma.2016.07.0030378-2166/© 2016 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
J. Lindström, S. Karlsson / Journal of Pragmatics 106 (2016) 129--147130
patients’ claims of no-knowledge in their responses to questions in medical interviews where patients are seeking relief fortheir rheumatism and fibromyalgia. While it is in the patients’ best interest to co-operate with the attending physician, thedata display frequent claims of no-knowledge on the part of the patients. In fact, a response containing a claim of no-knowledge usually reflects some knowledge rather than a lack thereof (compare Beach and Metzger, 1997; Tsui, 1991).Specifically, we explore question/answer-sequences where speakers, when answering, claim no-knowledge but whononetheless subsequently provide a candidate answer to the question. We argue that what is apparently logical paradoxis in fact a resource for speakers to signal interactional problems that are contingent on epistemic asymmetries betweenthe participants. In short, claiming no-knowledge becomes a resource for dealing with the tension between the biomedical,professional knowledge and the personal, lay experience of health issues.
This is indicated by the fact that no-knowledge claims are by their very nature nonconforming to type-specifyingquestions. These include questions that request a certain type of reply: interrogatives request a yes/no-answer, WH-questions request places, times, reasons, circumstances, or individuals, etc. (compare Heritage, 2012; Heritage andRaymond, 2005; Raymond, 2003). Extract (1) serves as an illustration: the doctor is inquiring about the patient’s generalwell-being, aside from the strictly physical experience of the illness, and has moved on to the patient’s psychological state.The extract begins with the doctor asking whether the patient is prone to depression, in other words, this is a question thatis based on the doctor’s professional knowledge; he has made the assumption that depression might be a side-effect ofthe illness.
(1) I NK: 8. D=doctor, P=pa tient.
01 D: depressiv,
‘depressive,’
02 (0.8)
03 P: #n:ä# ja vet int-
no I know not
‘no I don’t know-’
04 (0.3)
05 P: .hh
06 (0.4)
07 P: ehhh
08 (0.6)
09 P: m: man blir ju- .hh
PRT GNR becomes PR T
‘hm you do get’
10 .hh int vet ja ↓depressiv men ↓ #e[:# ]
not know I depressive but
‘I don’t know depressive but erm’
11 D: [.h ]
J. Lindström, S. Karlsson / Journal of Pragmatics 106 (2016) 129--147 131
15 s[å,#]
‘so’
16 D: [m ] mm:m
17 P: #e# no blir man [ju lite] så här
PRT PRT becomes GNR PR T a.little DE M
‘um sure you do become a little sort of’
18 D: [mm. ]
19 (1.1)
20 D: °preci[s.° ]
‘right’
21 P: [att in]t e man ju p[å- ]
that not is GNR PRT on
‘so you’re not’
22 D: [ °n]ä°
‘no’
23 P: på *gott humör* pre[cis int.
on good mood precisely not
‘exactly in a good mood’
24 D: [mm.
12 P: no blir man ju lite så hä
PRT becomes GNR PR T a.little DE M
‘sure you do get a little like’
13 irriterad om man int sover ord ↑entlit å?
irritated if GN R not speeps properly and
‘annoyed if you don’t sleep properly and’
14 #å# om man har ont #hela tiden
and if GN R has pain whole time-DE F
‘and if there is constant pain’
Although the doctor’s question in line 1 is highly elliptic, it is type-specifying, and requests the specific response type of a yes/no-answer. The patient’s reply in line 3 suggests that supplying this type of answer to the question is somehow problematic: nä
ja vet inte ‘no I don’t know.’ While the no-preface grammatically conforms to the yes/no-interrogative, the ‘I don’t know’
J. Lindström, S. Karlsson / Journal of Pragmatics 106 (2016) 129--147132
suggests something less definitive than the type-conforming ‘no,’ as it circumvents the sequential consequences of what ayes or a no would lead to (compare Raymond, 2003:949). This disclaiming of knowledge is interrupted in line 3 and is notconsidered to be a sufficient response by the doctor who does not take the turn at this point. After hesitating slightly, the patientresumes in line 9, pauses again, and inserts a knowledge disclaimer before producing the word depressive ‘depressive’ in line10. The formulation brings this word into question, which is introduced in the doctor’s initiative. The patient then continuestalking and explains how one’s physical condition and sleep deprivation affect one’s mood, concluding that int e man ju pågott humör int ‘it’s not like one’s in a good mood, right’ in lines 21 and 23. The doctor--patient negotiation on the correct labelfor the patient’s experience manifests itself as a clash between the participants’ differing access to relevant knowledge. Thisrefers to the doctor’s professional experience, which makes a question about depression relevant, and the patient’spersonal access to her experiences and feelings. The markers of no-knowledge therefore highlight the presuppositions ofthe question. In other words, the patient is not willing to accept the full consequence of agreeing with the description‘depressive,’ which is not only a psychological state, but also a choice of terminology that turns the psychological state intosomething pathological, treatable, and, in some societies, something rather shameful. As it transpires, due to the sideeffects of her illness the patient is not feeling entirely well, but she is unwilling to concede to the doctor’s proposed diagnosis.
Previous research on epistemic disclaimers has focused on their role in moves that are dispreferred, uncertain, andresisting (see Lindström et al., 2016). These uses are characterized as pragmatic or interactive, to distinguish them from theliteral (or cognitive) meaning of ‘I don’t know’ as a straightforward stand-alone response to an information-seeking question.In our present study, our main objective is to expand on previous linguistic knowledge of epistemic disclaimers by exploringtheir use in medical interviews, concentrating on how the specific circumstances of medical interaction provide fertilegrounds for the use of these expressions. Hence, this study also adds to the vast amount of conversation analytic researchon medical interviews (compare Maynard and Heritage, 2005). We will particularly relate to the participants’ epistemicasymmetries and how the patients cope with these asymmetries by claiming no-knowledge. To address this, we focus onthe sequences that involve patients marking their personal, subjective position in their interaction with medicalprofessionals who, nonetheless, possess superior biomedical knowledge of the matters discussed during consultations.The analysis therefore contributes to the discussion of epistemic dimensions in conversations, particularly thoseconcerning access and primacy (see Stivers et al., 2011). This analysis also adds to the research conducted on epistemicdisclaimers (compare among many others Degand et al., 2013; Tsui, 1991; Weatherall, 2011; Ӧstman, 1981) in that thedata include two varieties of Swedish: Sweden-Swedish and Finland-Swedish (see Norrby et al., 2012). Parallel to the otherstudies in this collection, the present analysis will thus contribute to generalizability across languages.
The following section will present a brief orientation of basic epistemic asymmetries in medical interaction withparticular emphasis on the dimensions of epistemic access, primacy, and responsibility. Section 3 introduces the data andthe principles for the collection of actual cases for analysis, that is, turn-prefacing epistemic disclaimers. Section 4presents our core analysis of epistemic disclaimers as a resource for dealing with the participants’ different positions inrelation to epistemic access and epistemic rights. The article closes with conclusions drawn from the analysis (Section 5).
2. Background: epistemic asymmetries in medical interaction
On a general level, the doctor and the patient share the same goal during a medical consultation in that the patientshould receive help and be cured. Medical interaction is nonetheless characterized by several asymmetries that arecaused by the differences in the participant roles, knowledge bases, and expectations (see Drew, 1991). While the doctoris the expert who possesses formal biomedical knowledge and authority over health and medicine, the patient has a socialand personal lay perspective on such issues. This means that doctors orient to the general, objective knowledge andprofessional practices in their field, which may clash with patients’ particular and subjective experience of their symptomsand ailments. As an expert, the doctor is responsible for an agenda that is relevant to the patient’s needs, whereas thepatient is entitled to receive relevant information and treatment.
With regard to claims of no-knowledge, the key interactional issue that arises is the asymmetries in the orientations to theepistemic domain. To make an objective conclusion and diagnosis, the doctor requires specific and often measurableinformation, such as the duration, location, and intensity of the patient’s pain. However, the patient may have difficulty inproviding this type of information because the subjective experience of pain is acute and prevalent. As a consequence, it maybe demanding to describe pain in exact terms if the patient has not made notes of the times and the physical symptomsinvolved. We argue that this type of problem was illustrated by extract (1): the doctor was seeking a specific label for thepatient’s mental state, whereas the patient expressed trouble in finding one word that would capture and specify her situation.
With reference to Stivers et al. (2011), we find it analytically beneficial to relate epistemic asymmetries in medicalconsultations to the dimensions of access, primacy, and responsibility, as illustrated in Table 1.
As regards epistemic access, the patients have direct, first-hand knowledge of their own physical experience, whereasthe doctors have superior general knowledge of medical symptoms and their causes. When we consider epistemicprimacy, the patients are entitled to their subjective experience as well as their right to communicate it. The doctors instead
J. Lindström, S. Karlsson / Journal of Pragmatics 106 (2016) 129--147 133
Table 1Epistemic dimensions and asymmetries in medical interaction.
Dimension/participant Epistemic access Epistemic primacy Epistemic responsibility
Patient First-hand knowledge of one’sphysical/mental experience
Right to describe one’s physical/mental experience
Provides adequate information onone’s physical/mental experience
Doctor Professional knowledge based onmedical education and experience
Authority to decide what is relevantand to diagnose and prescribe
Provides adequate medicalinformation and treatment
have the authority to decide what is relevant to discuss, diagnose, and suggest as the patient’s treatment. The doctors areresponsible for giving adequate diagnoses, information, and treatments, but this task depends on intersubjective trust.This means that the information that the patients provide likewise needs to be adequate and truthful. Indeed, the doctorsmay be held accountable for withholding requested information, which we illustrate in extract (2): the doctor claims a lackof knowledge with his stand-alone disclaimer as an answer to the patient’s question in line 4:
(2) INK:12.
01 P: >va gör he de lymfaterapin (.)
what does DE M lymphatic-therary-DE F
‘what does it do, the lymphatic therapy (.)‘
02 va händer i kroppen<
what happens in body-DE F
‘what does it do to the body’
03 (.)
04 D: ja vet int.
I know not
‘I don’t know’
05 (.)
06 P: du vet int he.
you know not that
‘you don’t know that’
07 D: fråga henne.
‘ask her’
08 (.)
09 D: *(nä) de vet ja int faktist.*
no that know I not really
‘(no) I actually don’t know that’
The patient is interested in testing lymphatic therapy and the doctor has encouraged her to pursue that. In line 1, thepatient inquires as to how the therapy works and the doctor replies to this, ja vet int ‘I don’t know.’ This is followed by thepatient’s somewhat surprised receipt, du vet int he ‘you don’t know that.’ The doctor then refers to the physiotherapist
J. Lindström, S. Karlsson / Journal of Pragmatics 106 (2016) 129--147134
( fråga henne ‘ask her’) as a source of knowledge. In line 9, he continues on to restate, this time with a smiley voice, that hereally does not know how the therapy works -- the adverb faktiskt, ‘actually, factually’ implies that the content of what isbeing said runs counter to the other party’s expectations. These claims of no-knowledge by doctors are rare in our data,and the instantiation in (2) might be explained by the fact that lymphatic therapy is not a part of Western standard medicine.Hence, the doctor problematizes his responsibility to provide information on something that is assumed to be within hisarea of knowledge (see Keevallik, 2011:193). However, rather than signal an unwillingness to answer, he claims no-knowledge. Not being able to contribute leaves the doctor less culpable than if he had appeared unwilling to supplyinformation about the treatment (compare Clayman, 2002; Heritage, 1988).
Doctors may be equally surprised when patients claim a lack of knowledge in matters that they alone are expected tohave primary epistemic access to. Extract (3) illustrates this type of orientation: the doctor has asked whether thesymptoms that the patient has described are something that her parents have suffered from, and the patient hasresponded by talking about her elderly mother. We enter the transcript when the doctor poses a specifying question(appended to an earlier patient turn which is not seen here): när hon va i samma ålder ‘when she [the patient’s mother] wasat the same age [as the patient herself at the time of speaking]?’:
(3) INK:12.
01 D: när hon va i samma ålder¿
when she was in same age
‘when she was the same age¿’
02 (.)
03 P: ja veit int.
I know not
‘I don’t know’
04 D: aj [du vet int.
PRT you know not
‘oh [you don’t know’
05 P: [mamm ha vari arbetsam å gai (ja)
mom has been hard.working and gone PR T
‘mom’s been hard working and walking (well)’
06 [nu allti å ( ) gör,
PRT always and does
‘all the time and ( ) does,’
07 D: [hon ha inte prata om de.
she has not talked on it
‘she hasn’t talked about it.’
08 (.)
09 P: nä:ä
‘no’
J. Lindström, S. Karlsson / Journal of Pragmatics 106 (2016) 129--147 135
In line 3, the patient claims to lack knowledge by using the regional form ja veit int, which is an utterance that hasthe prosodic features of a completed reply, and the doctor responds to this as something unexpected. His turn inline 4 is initiated by his change of state token aj ‘oh,’ which is typical of Finland Swedish (see Lindström and Wide,2015) and interprets the ‘I don’t know’ response as full-fledged in the sequential context (although the patientcontinues in line 5). Family history, including diseases and ailments, is potentially something that the patient can haveaccess to but otherwise, her mother’s experiences would have only been known to her if the mother had discussedthem with her. This type of knowledge is therefore occasioned (Keevallik, 2011:196; Pomerantz, 1980:187). Thedoctor draws the appropriate conclusion and acknowledges the evidential background for the patient’s lack ofknowledge in line 7.
These types of epistemic landscapes tend to create tensions between the doctors’ and patients’ orientations toknowledge. In the following, we focus specifically on the turn or TCU-prefacing jag vet inte ‘I don’t know’ (with some formalvariants), with the aim of specifying what these epistemic tensions entail and how they are dealt with in the interactionalwork displayed by the participants.
3. Data and collection
The data consist of two corpora of medical interactions, one where the interactants speak Sweden-Swedish varietiesand the other where they speak Finland-Swedish varieties. The Sweden-Swedish data (LOP) were collected at arheumatology clinic and consist of 15 consultations for a total length of 7 h. The interactions include both initial visits, whenthe patient enrolls at the clinic, as well as visits by more experienced patients who have been treated for some time(Melander Marttala, 1995). The Finland-Swedish data (INK) are from a similar environment and also include both initialvisits and follow-up visits, amounting to 20 consultations with a total length of 13 h (Lindholm, 2003). The patients in thedata from Finland suffer from fibromyalgia.1 Both sets of data concentrate on the patients’ experiences of pain and therelated treatment, and the interaction between the doctors and patients focuses on the patient’s experience andrecollection as well as the doctor’s history-taking. A physical examination also occurs during the consultations. The datawere video recorded and collected with the informed consent of all participants and was subsequently transcribed. Theextracts have an idiomatic English translation. A word-by-word gloss is provided when the word order in the translationdiffers from the original.
From this data, we have discovered 29 excerpts of variations in the Swedish epistemic disclaimer jag vet inte ‘Idon’t know,’ that occur within patient turns (compare Dutch ik weet niet, German ich weiss nicht; compare also Helmeret al., 2016). Most of these instances have the straight word order jag vet inte, but some examples contain an invertedword order with the negation first (int vet jag), or with an anaphoric pronoun (det ‘that’) in the initial position (det vet jaginte). These variations occur due to the structural properties of Swedish and all of them have a routinized status. Thecommon principle for the excerpts was to include uses with the epistemic disclaimer constituting some form ofprefatory move in patient response turns, either as a separate turn constructional unit, or as a preface to a turnconstructional unit (TCU); (compare Schegloff, 1996; Weatherall, 2011). We are therefore interested in the pragmaticpremodifying properties of the knowledge disclaimer within the realm of a conversational turn. The occurrences instand-alone answers were excluded, because these differ from the uses of the epistemic disclaimer as a turn-framingdevice; for a discussion of the differences, see Keevallik (2011) on Estonian. The meaning of jag vet inte in stand-alone answers may be literal or remain implicit until a further course of interaction unfolds. We have also not includedinstances of jag vet inte that were followed by a complement clause, such as jag vet inte om det är sant ‘I don’t knowwhether it is true,’ jag vet inte hur det är möjligt ‘I don’t know how it is possible.’ In these cases, jag vet inte is not apreface because the meaning is literal and its target specific, which is expressed in the subsequent dependentclause.
The three forms of the epistemic disclaimer -- jag vet inte, inte vet jag, and det vet jag inte-have different distributions inthe data. The first variant, with the subject first, is clearly the most frequent and general form, whereas the second variant(with negation first) only occurs in the consultations recorded in Finland (compare Laury and Helasvuo, 2016, onnegation-first disclaimers in Finnish). However, this variant is not as frequent as has been reported in everyday Finland-Swedish conversation. A study of Helsinki Swedish teenage talk reported the proportion between jag vet inte and inte vetjag to be approximately 50/50 (J. Lindström, 2009), whereas the ratio in the INK corpus is 29/5 (note that stand-alonedisclaimers were included in these general counts, but not in our collection of excerpts). One reason for the low
1 Fibromyalgia is a chronic disease that is characterized by widespread pain, fatigue, and joint stiffness. As the data from Swedenwere collected at a rheumatology clinic, we could say that the ailments the patients suffer from appear to be sufficiently similar in both datasets.
J. Lindström, S. Karlsson / Journal of Pragmatics 106 (2016) 129--147136
frequency of the negation-first disclaimers can be their general pragmatic signal value: they commonly appear in turnsthat by nature are overtly resisting or rejecting (J. Lindström, 2014), as we can also see in extract (1), line 10. This mightbe the reason that patients avoid sending such interactional signals during medical interviews, which are basically co-operative in nature.
Finally, the variant det vet jag inte as a prefatory move occurs in the data only three times, and owing to this lowfrequency, it is not possible to claim anything decisive about its pragmatic meaning. Nonetheless, these instancesoccur in the Sweden-Swedish dataset and they occur when a patient is confronted with a question that exceeds thelimits of his or her competence, that is, there is an issue with epistemic access. An example of this use is shown inextract (4); at this point of the transcript, the patient admits to having ingested non-prescription painkillers before theconsultation.
(4) LOP:3.
01 P: nu fuska ja fak[tist] å
now cheated I really and
‘I actually cheated now and’
02 D: [jaa ]
’yeah’
03 P: tog #e- en en Magnecyl inn[an ] ja gick
took a a Magnecyl before I went
‘took an- an an Aspirin before I left’
04 D: [°.ah°]
05 P: för att ja [skulle kunna ta-#
because I would be.able take
‘to be able to take-’
06 D: [hjälper hjälper en Magnecyl,
helps helps a Magnecyl
‘does does an Aspirin help,’
07 P: de #vet# ja #in#te. (.)
that know I not
‘I don’t know that’
08 °men ja kan ju inbilla mej att de [gör] de.°
but I can PRT imagine RF L that it does it
‘but I can tell myself that it does.’
09 [mm ]
As the patient takes painkillers before a medical consultation, this could affect the outcome of the consultation, and this issomething that is suggested by the patient’s choice of words in line 1, nu fuska ja faktist ‘actually I cheated this time.’
J. Lindström, S. Karlsson / Journal of Pragmatics 106 (2016) 129--147 137
However, the doctor does not reproach her in line 6, but instead asks if the substance has had any effect; formulatedin this way, the question puts the patient’s own experience of the physical effects at stake. In line 7, the patientsubsequently claims no-knowledge, de vet ja inte ‘that I don’t know,’ and continues on to suggest that thesupposed effect might be imaginary. A ‘yes’ or ‘no’ response here, which would match the form of the question,would be inadequate, and would be over-reaching as to what she can actually know, or claim to know, aboutthe effects of the drug. The topicalization of the object (det ‘that’) in this variant of epistemic disclaimer may indicatethat the disclaimer is more specific than the variants that lack the anaphoric pronoun; in other words, the speaker isexpressly distancing herself from an element in the previous turn (compare Karlsson, 2006:143--144).
4. Analysis: dealing with epistemic access and primacy
In the following subsections, we analyze the patients’ use of turn-prefacing or TCU-prefacing epistemic disclaimers.In the former case, the epistemic disclaimer constitutes a TCU of its own, but in both cases, it is a prefatory interactionalmove that is followed by more talk by the same speaker (compare Weatherall, 2011). The analysis accounts forepistemic asymmetries that locally occur in question--answer sequences during doctor--patient interactions. By focusingon the dynamics between the doctors’ questions, the patients’ answers, and the participants’ epistemic statuses (seeSection 2 above), we identify two types of contexts that produce turn/TCU-framing ‘I don’t knows’. The first is whenparticipants are dealing with differing epistemic access and the second is when participants’ epistemic rights (primacy)are (re)negotiated.
4.1. Differences in epistemic access: heading off an inadequate assumption
Extract (4) is an example of problems with epistemic access, and possibly also with epistemic primacy in the medicaldomain, which were oriented to by using the negative epistemic marker. In the following, we discuss instances of thetension between the speakers’ access domain surfacing even more clearly. The doctors’ questions during patienthistory-taking inevitably contain propositions that presuppose circumstances about the illness and the patient’sexperiences of the illness. In effect, by posing a question in a certain manner, the doctor may preframe a certain type ofanswer. When the doctors use their professional judgements to diagnose patients, the assumptions underlying theirquestions often stem what they deem relevant to describe. As was seen in extract (1), these types of questions maysuggest descriptive labels that the patient cannot accept because they contradict with what the patient knows about herown condition. Extract (5) is a further example; the physician is enquiring about the character and location of the pain thepatient is experiencing:
(5) LOP:12.
01 D: har du nå ont i (.) nacken eller=
have you some pain in neck-DEF or
‘do you have any pain in (.) your neck or’
02 P: .nähh: ingenting sånt.
no nothing such
‘no nothing like that’
03 D: ont i ryggen eller [°höften¿°]
pain in back-DE F or hip-DE F
‘pain in your back or hip’
04 P: [nä: ja ↑rygg ↓en .] (.)
no yes back-DE F
‘no yes my back (.)’
J. Lindström, S. Karlsson / Journal of Pragmatics 106 (2016) 129--147138
PRT PR T I know not I have not pain
‘yeah, see, I don’t know I’m not in pain’
06 men de känns:=ja blir trött=
but it feels I become tired
‘but it feels: I get tired’
07 =men de kan ju va kondition som=
but it can PR T be condition that
‘but that could be my bad shape that-’
08 =ja e otränad igen nu.
I am untrained again now
‘I’m unfit again now.’
09 D: mm:m¿
10 P: så de känns:? (.)
‘so it feels’
11 de e inte värk men #eh#- (.)
it is not pain but
‘it’s not pain but um-’
12 de e (.) länd#ryggen liksom#,
it is lower.back-DE F like
‘it’s like (.) my lower back right,’
13 D: mm:m
05 a se ja vet inte ja har inte ont
In line 1, the doctor asks the patient if she is experiencing pain in her neck, which she denies. The doctor then continues byadding two further candidates for the location of her pain: either her back or hip. The patient’s initial response is negative,but she immediately reacts to the physical location ryggen ‘the back’ in line 4. However, in line 5, the patient problematizesthe question: ja vet inte ja har inte ont ‘I don’t know I’m not in pain,’ followed by three lines, 6--8, that constitute attempts tonarrow down the sensation that corresponds to something closer to her own experience than ‘pain.’ The doctor only givesa minimal response, and the patient continues to specify the problem. The question presupposes that pain is a relevantfocus, something that patients would indeed normally see treatment for at a rheumatology clinic. For the patient, however,‘pain’ is not the correct word to describe her problems, even though she has had some problems with her lower back. Thedouble proposition of the question, that there is pain and that it could be located in the patient’s back, makes it difficult toreply to when both propositions are not relevant to the patient (compare Linell et al., 2003).
Extract (6) contains a similar defiance of an assumption that has been embedded in the question. In line 1, the doctorrequests information by using a declarative statement concerning circumstances that are clearly within the patient’sepistemic domain (compare Labov and Fanshel, 1977). The format of the question suggests that the doctor is relativelycertain of his proposition; the statement is something of a conclusion based on a patient’s previous statement regardinghow good it is to have people to talk to about her ailment.
J. Lindström, S. Karlsson / Journal of Pragmatics 106 (2016) 129--147 139
(6) INK:8.
01 D: ni har en ganska aktiv förening där=
you. PL have a rather active club there
‘you’ve got a rather active club there’
02 P: =ne ja vet int vi- .h alltså Irina å ja
no I know not we PRT Irina and I
‘nah I don’t know we- .h that is Irina and I’
03 ha egentligen jo- (0.3) jobba tisammans [förut] att
have actually worked together before that
‘have actually wo- worked together before so’
04 D: [mt ]
05 D: precis j[å ]
‘right yes’
06 P: [att] vi ha int egentligen vari så där aktiva
that we have not actually been DEM active- PL
‘so we haven’t actually been all that active’
07 den e ju ganska n[y den hä [föreningen
it is PRT fairly new DE M club-DE F
‘it’s fairly new this club, you know’
08 D: [jå [jå
‘yes yes’
09 P: [i ((Lillköp[ing)) att vi ha int
in Lillköping that we have not
‘in Lillköping so we haven’t’
10 D: [jå [.ju
‘yes right’
11 (0.2)
12 D: jåjå
yes-yes
‘yeah yeah’
J. Lindström, S. Karlsson / Journal of Pragmatics 106 (2016) 129--147140
13 (0.3)
14 P: ha int egentligen vari så hemst ak[tiva i den]
have not actually been so awful active-PL in it
‘haven’t actually been all that active there’
The patient’s response in line 2, ne ja vet int ‘nah I don’t know,’ suggests that although the doctor bases his statementon information provided by a person with epistemic access and whose epistemic domain the issue belongs to(Keevallik, 2011), there is something about his proposition that the patient cannot accept without modification. Theproposition suggested in the question is, on the one hand, that the club is active, and on the other, that the club (and itslevel of activity) is relevant to the patient’s reported feeling of relief about having people in the same situation to talk to.The patient proceeds to explain how the doctor’s conclusion focuses on the wrong point -- the support she wasreferring to hinges not so much on the club as it does on one of the members being a previous co-worker. This isfurther elaborated in lines 6--9 and 14, where the doctor’s question is explicitly challenged; the club was formedrecently and has not yet organized many activities. By resisting the format of the question and the doctor’s claim to beable to accurately reformulate the patient’s information, she has an opportunity to set the record straight. In otherwords, she reclaims her epistemic primacy, asserting herself as the primary interpreter of the matters within herepistemic domain.
Both extracts (5) and (6) are examples of the patient utilizing a statement of no-knowledge to correct an assumption oran interpretation of something that is not within the questioner’s primary epistemic reach. In (5), the doctors’ questionssuggest physical and psychological experiences that might be relevant for the patient to accept or reject in a yes/no-polarity format. A question formed as a declarative (6) likewise creates an opportunity to respond in a yes/no-polarityformat. In both extracts, the patients reject the type-specifying format and instead accept only part of the underlyingpropositions of the question, presenting a somewhat more nuanced description of what they accept as relevant. Anepistemic disclaimer serves as a bridge for nuanced formulation, which signals that one is distancing oneself from whathas been said previously.
4.2. Negotiating epistemic rights by taking a more independent stand
The above extracts have illustrated sequences that contain epistemic asymmetries leading to interactive failure;the doctors posed questions based on presuppositions that conflicted with the patient’s primary knowledge. Closelyintertwined with these types of interactionally problematic situations is the question of the participants’ epistemicprimacy -- who has the right to (adequately) know and what. This was also evident in extract (6). In the following, wewill discuss further instances in which epistemic rights are at stake and in which the patient adopts a moreindependent stand in relation to the doctor’s a priori epistemic authority. Extract (7) serves as a mild and embeddedexample of these types of interactional tensions, which resembles extract (5), as here the doctor asks the patientwhether she can localize her pain, suggesting her arms, legs, and back as possible locations. In other words, hemakes assumptions based on his medical experience. The patient initiates a response by orienting to the doctor’ssuggestions and confirming them. In line 6, however, she produces ja vet int ‘I don’t know,’ which marks a shift in herepistemic posture.
(7) INK:10.
01 D: nå dedär: ä (.) kan du säga att du ha värkt (.)
PRT PRT PR T can you say that you have ached
‘right um (.) would you say that there is pain (.)’
02 i (.) överarmar å i ben å (.) i underarmar (.)
in upperarms and in legs and in forearms
‘in (.) your upper arms and your legs and (.) forearms’
J. Lindström, S. Karlsson / Journal of Pragmatics 106 (2016) 129--147 141
03 [i ryggen,
in back-DEF
‘in your back’
04 P: [i ry- i ryggen a ha [också (.) [mm (.) å i benen också [i i
in in back-DE F PR T also PR T and in legs-DE F also in in
‘in my back uhhuh also (.) mm (.) and in my legs too in in’
05 D: [i ryggen [jå [mm
in back-DEF yes PR T
‘in your back, right, mm’
06 P: ja vet int (.) ibland s- (.) tycke ja bara att ja inbilla me
I know not occasionally think I just that I imagine RF L
‘I don’t know (.) sometimes (.) I just think that I’m imagining’
07 att [va ha ja fö fel att (.) att
that what have I for fault that that
‘like what’s wrong with me (.) so that’
08 D: [mm
09 P: de kan ju int va sån- i- int kan ja ju va sjuk genom
it can PRT not be such not can I PR T be sick through
‘it can’t be like this- I can’t be sick all through’
10 hela [kroppen int.
whole body-DE F not
‘my body, right.’
11 D: [mm
12 (.)
13 D: nå de e å möjlit fös*tås*.
PRT it is also possible of.course
‘well that’s possible too of course.’
The contribution that follows the claim of no-knowledge is framed as an incidental subjective experience, referring to the‘occasionally I just think that I’m imagining’ in line 6. This latter part of the patient’s turn is thus devoted to communicating apersonal, more independent perspective, which is clinically less exact but it is apparently subjectively more accuratebecause she might feel pain throughout her whole body. The hedged and negatively framed formulations in the latter partof the patient’s turn (int kan ja ju va sjuk genom hela kroppen int ‘I can’t be sick all through my body, right,’ lines 9--10 and13) indicate that claiming epistemic primacy is a socially delicate issue in a context that involves the doctor’s domain in theinstitutional order (Gill et al., 2001). The epistemic disclaimer marks this dispreferred shift in posture from an initial
J. Lindström, S. Karlsson / Journal of Pragmatics 106 (2016) 129--147142
concessive stance to her making a suggestion, or even devising her own hypothesis. At the end of the extract in line 13,the doctor acknowledges the patient’s sentiments and the relevance of her description with a smiley voice.
Epistemic primacy becomes an even clearer issue in extract (8). The patient has reported that she is taking rather largedoses of a certain (mild) medication (‘ten capsules a day’) and that this is expensive. In line 1, the doctor states that theamount seems large and poses a challenging question (behöver du så mycke ‘do you need that much?’). The patientreplies in line 4 in a turn prefaced with the particle nå (A. Lindström, 2016), which may signal a dispreferred, or at least anon-straightforward, answer, followed by ja vet int. As in (7), the turn continues with a mitigating incidental description(ibland sen så ‘occasionally then’), but this line of reasoning is aborted; instead, the patient continues by declaring thatsome days she does not take the medication at all, which can be interpreted as a defensive move in lines 5--6.
(8) INK:13.
01 D: de låter ganska mycke [de jå
that sounds quite much that yes
‘that sounds like quite a lot, yes’
02 P: [ja
‘yes’
03 D: ö: behöver du så mycke (.) [( )
PRT need you so much
‘um do you need that much (.)’
04 P: [nå ja vet int ibland sen så,
PRT I know not occasionally then PR T
‘well I don’t know sometimes then,’
05 nå nu ha ja int kanske tagi igen (.) på nån,
PR T now have I not maybe taken again on some
‘well, now I haven’t had any again (.) for,’
06 några dagar ha ja vari utan [helt.
some- PL days have I been without wholly
‘for a few days I haven’t taken any at all.’
07 D: [mm:m
08 (0.4)
09 P: men nu tar ja då när ja känner att att ja behöver=
but PRT take I then when I feel that that I need
‘but I do take some when I feel the need’
10 =[egentligen tycker ja att ja känner de på mej.
actually think I that I feel it on me
‘actually I think I can feel it.’
J. Lindström, S. Karlsson / Journal of Pragmatics 106 (2016) 129--147 143
that I can have pauses PR T
‘so I can have breaks definitely’
14 D: jå:å
yeah
‘right’
15 P: .jå
‘yeah’
16 D: jå:å
yeah
‘right’
17 P: °för att int e de ju bra heller°
because not is it PRT good either
‘°because it’s certainly not good either°’
18 (0.3)
19 för levern så här att man tar massor me [.h
for liver-DEF like this that GN R takes masses with
‘for the liver that you take a bunch of’
20 D: [mm
21 P: olika saker.
different things
‘diverse stuff’
22 D: ja å om int annat så blir de som sagt dyrt.
yes and if not else PR T becomes it as said dear
‘yes, and if nothing else, as (you) said, it gets expensive.’
11 D: [mm:m
12 D: oke[j
‘okay’
13 P: [att ja kan ha pauser #nog#.
In line 9, following a short pause, the patient adopts a more definite stand: she states that she takes the medicationwhen she feels that she needs it, underlining in line 10 that she is able to identify and regulate her own needs. Byclaiming this, the patient assumes a noticeable independent epistemic posture, claiming epistemic primacy in themedical issue being discussed. The turn-initial epistemic disclaimer in line 4 can therefore be analyzed as a prefatory
J. Lindström, S. Karlsson / Journal of Pragmatics 106 (2016) 129--147144
marker. It prefaces the distance-taking from the physician’s challenging question and the epistemic primacyincorporated in it, foreshadowing the patient’s shift to an independent stance in lines 9--10. The doctor respondswith a series of continuers, leaving the turn for the patient, who in lines 17--21 concedes that large doses can damagethe liver. This concession may be an orientation to pre-empt a possible critical professional stance from the physician,but it also displays that the patient is appropriately oriented to lay versus professional knowledge in relation to healthissues (compare Gill et al., 2001). The physician does not reclaim epistemic primacy in that sequence, but onlyreiterates (line 22) what the patient initially brought up (note the phrase som sagt ‘as was/you said’): that taking largequantities of medicine is (at the very least) an economic issue.
Extracts (7) and (8) both illustrate tensions that arise due to the participants’ orientations to epistemic primacy andtheir shifts in epistemic posture as a result of the tensions. The institutional frame provides the doctor with epistemicprimacy and, consequently, when the patient takes the initiative in the medical domain, it is a delicate matter. Thesetypes of initiatives are therefore surrounded by signals of their socially dispreferred nature: the patient may producepartly conforming or conceding answers and frame her independent stand with hedging devices. Together with otherhedging devices, the epistemic disclaimer signals that the patient is reluctant to accept the line of reasoningsuggested by the other; instead, the patient is heading to a more independent epistemic position (compare Heritage,2002). In doing this, the patient asserts more agency over her response (Stivers and Hayashi, 2010; Stivers et al.,2011:22), which may enable her, for example, to claim health-monitoring competence, as is clearly the case in (8)(see Gill et al., 2001).
5. Conclusions
The present analysis points to interactional practices used to assert one’s relative epistemic positioning (Heritage,2012; Heritage and Raymond, 2012). The assumptions of the questioner concerning the knowledgeability of theconversational partner can be adjusted in the response turn by expressing that one has no knowledge of the matter beingdiscussed. We therefore argue that explicitly claiming no-knowledge is not the same as providing a non-answer response(compare Stivers and Hayashi, 2010:23). Instead, such responses resist something about the question (compareKeevallik, 2011). By studying response turns that are framed by the epistemic disclaimer jag vet inte, we have discernedcertain interactional organizing patterns. These types of responses are superficially anomalous: the speaker first deniesknowledge but then, and in the continuation of the response, still maintains having some type of knowledge. Most of theprevious studies on turn-prefacing epistemic disclaimers demonstrate that they signal the speaker’s uncertainty regardingwhat is being said and that the epistemic disclaimer therefore serves as a hedge (see Lindström et al., in 2016). In thepresent study, we have extended this further in determining what could constitute these markings of uncertainty in thecontext of medical interaction.
One common feature of responses prefaced with an epistemic disclaimer is that the speaker (patient) can flag ananswer as less straightforward so that the assumptions embedded in the question may be faulty, conflicting with what thepatient has access to, or the question may infringe on the recipient’s (experienced) epistemic rights. The epistemicdisclaimer thus marks an interactional problem at hand and the speaker’s hesitancy regarding how to proceed with asocially delicate or complicated upcoming contribution. By claiming superior epistemic access, the patient may appear tobe more knowledgeable than the doctor, and by claiming epistemic rights, the patient may take initiatives that areconventionally reserved for the medical professional (for example, suggesting diagnosis or medication). Thus, the sourceof tensions in these epistemic dimensions is the lay--professional relation, and it is through epistemic disclaimers that thespeakers can mark their awareness of such tensions and navigate among them.
The prefacing epistemic disclaimer as a unit can be analyzed as a pragmatic marker that simultaneously linksbackwards and points forwards in discourse. The preceding move is marked as being problematic by ‘I don’t know’, but italso signals that the talk within its scope is somehow complicated. In other words, the disclaimer projects a modification ofthe proposition put forward by the questioner. In the examples analyzed in this study, the patient is constructed as havingepistemic access to certain domains. Answering with a no-knowledge claim involves re-positioning the speaker fromknowledgeable to non-knowledgeable, or rather, to knowledgeable of something other than what the question requested.This is how the speaker claims more agency over her response and works toward resolving the epistemic tension that hassurfaced in the interactional micro-situation.
Acknowledgements
We would like to thank Camilla Lindholm and Ulla Melander Marttala for their permission to use their data on medicalinteraction, the corpora INK and LOP, respectively. We would also like to extend our appreciation to the participants of twoconference panels devoted to negative epistemics in which the current study (among many others) was presented anddiscussed, The 13th International Pragmatics Conference in New Delhi (2013) and The 4th International Conference on
J. Lindström, S. Karlsson / Journal of Pragmatics 106 (2016) 129--147 145
Conversation Analysis at UCLA (2014). Finally, we would like to express our gratitude to the reviewers for their valuablecomments on earlier versions of the article, and to Kate Moore for checking the language. We are solely responsible forany remaining errors or shortcomings. Jan Lindström’s research was supported by Riksbankens Jubileumsfond (TheBank of Sweden Tercentiary Foundation) in the project Interaction and Variation in Pluricentric Languages --Communicative Patterns in Sweden Swedish and Finland Swedish (grant ID: M12-0137:1).
Appendix
Transcription symbols
. falling intonation
, level in tonation?
slightly
rising intonation ? rising i ntonation - cut off [ start of overlapping talk ] point w here overlapping talk stops (.) microp ause, less than 0.2 s (0.5) silence s timed in tenths of a second = ‘‘latchin g’’, i.e. no silence between two adjacent utterances > < talk ins ide is at a faster pace than the surrounding talk : lengthe ning of a sound mm:m stretch ed pronunciation (legato) en- cut off en empha sis indicated by underlining *en* smile v oice #en# creaky voice 8en8 talk ins ide is more quiet than the surrounding talk (en) uncerta in transcription ( ) inaudib le words .en word p ronounced with initial inhalation .hh audible inhalation (the more h’s, the more aspiration) hh audible exhalation (length as above)Glossing abbreviations
DEF definite
DEM demon strative GNR generic 3rd person pronoun (cf. ‘one’) PL plural PRT (discou rse) particle RFL reflexiv e (pronoun)References
Beach, Wayne A., Metzger, Terri R., 1997. Claiming insufficient knowledge. Hum. Commun. Res. 23 (4), 562--588. http://dx.doi.org/10.1111/j.1468-2958.1997.tb00410.x
Clayman, Steven E., 2002. Sequence and solidarity. In: Lawler, E.J., Thye, S.R. (Eds.), Group Cohesion, Trust, and Solidarity, pp. 229--253.Degand, Liesbeth, Cornillie, Bert, Pietrandrea, Paola, 2013. Discourse Markers and Modal Particles: Categorization and Description. John
Benjamins Publishing Company, Amsterdam.Drew, Paul, 1991. Asymmetries of knowledge in conversational interactions. In: Marková, I., Foppa, K. (Eds.), Asymmetries in Dialogue. Barnes
and Noble Books, Savage, MD, pp. 29--48.
J. Lindström, S. Karlsson / Journal of Pragmatics 106 (2016) 129--147146
Gill, Virginia Teas, Halkowski, Timothy, Roberts, Felicia, 2001. Accomplishing a request without making one: a single case analysis of a primarycare visit. TEXT 21 (1/2), 55--81. http://dx.doi.org/10.1515/text.1.21.1-2.55
Helmer, Henrike, Reineke, Silke, Depperman, Arnulf, 2016. A range of uses of negative epistemic constructions in German: ‘ich weiss nicht’ as aresource for dispreferred actions. J. Pragmat. 106, 97--114.
Heritage, John, 1988. Explanations as accounts: a conversation analytic perspective. In: Antaki, C. (Ed.), Analyzing Lay Explanation: A CaseBook of Methods. Sage, London, pp. 127--144.
Heritage, John, 2002. Oh-prefaced responses to assessments: a method of modifying agreement/disagreement. In: Ford, C., Fox, B.A.,Thompson, S.A. (Eds.), The Language of Turn and Sequence. Oxford University Press, Oxford, pp. 196--224.
Heritage, John, 2007. Intersubjectivity and progressivity in person (and place) reference. In: Stivers, T., Enfield, N.J. (Eds.), Person Reference inInteraction: Linguistic, Cultural and Social Perspectives, pp. 255--280. http://dx.doi.org/10.1017/CBO9780511486746.012
Heritage, John, 2012. The epistemic engine: sequence organization and territories of knowledge. Res. Lang. Soc. Interact. 45 (1), 30--52. http://dx.doi.org/10.1080/08351813.2012.646685
Heritage, John, Raymond, Geoffrey, 2005. The terms of agreement: indexing epistemic authority and subordination in talk-in-interaction. Soc.Psychol. Q. 68 (1), 15--38. http://dx.doi.org/10.1177/019027250506800103
Heritage, John, Raymond, Geoffrey, 2012. Navigating epistemic landscapes: acquiescence, agency and resistance in responses to polarquestions. In: de Ruiter, J.P. (Ed.), Questions: Formal, Functional and Interactional Perspectives. Cambridge University Press, Cambridge,pp. 179--192. http://dx.doi.org/10.1017/CBO9781139045414.013
Hutchby, Ian, 2002. Resisting the incitement to talk in child counselling: aspects of the utterance ‘I don’t know’. Discourse Stud. 4 (2), 147--168.http://dx.doi.org/10.1177/14614456020040020201
Karlsson, Susanna, 2006. Positioneringsfraser i interaktion. Syntax, prosodi och funktion. Gothenburg University, Gothenburg.Keevallik, Leelo, 2011. The terms of not knowing. In: Stivers, T., Mondada, L., Steensig, J. (Eds.), The Morality of Knowledge in Conversation.
Cambridge University Press, Cambridge, pp. 184--206. http://dx.doi.org/10.1017/CBO9780511921674.009Labov, William, Fanshel, David, 1977. Therapeutic Discourse: Psychotherapy as Conversation. Academic Press, New York.Laury, Ritva, Helasvuo, Marja-Liisa, 2016. Disclaiming epistemic access with ‘know’ and ‘remember’ in Finnish. J. Pragmat. 106, 80--96.Lindholm, Camilla, 2003. Frågor i praktiken: flerledade frågeturer i läkare-patientsamtal. (diss.). Svenska litteratursällskapet i Finland, Helsinki.Lindström, Jan, 2009. Negationen inte i satsens spets -- int e de ju rimligt. In: Wide, C, Lyngfelt, B. (Eds.), Konstruktioner i finlandssvensk syntax.
Svenska litteratursällskapet i Finland, Helsinki, pp. 163--189.Lindström, Jan, 2014. Front field negation in spoken Swedish: a regional archaism? In: Taavitsainen, I., Jucker, A.H., Tuominen, J. (Eds.),
Diachronic Corpus Pragmatics. John Benjamins Publishing Company, Amsterdam, pp. 237--253. http://dx.doi.org/10.1075/pbns.243.14lin
Lindström, Anna, 2016. Nå in Swedish conversation. In: Auer, P., Maschler, Y. (Eds.), ‘‘Nu’’ and ‘‘Nå’’: A Family of Discourse Markers Across theLanguages of Europe and Beyond. Walter de Gruyter, Berlin.
Lindström, Jan, Wide, Camilla, 2015. Finlandssvenskt samtalsspråk. In: Tandefelt, M. (Ed.), Gruppspråk, samspråk, två språk: Svenskan i Finland-- I dag och i går. Svenska litteratursällskapet i Finland, Helsinki, pp. 91--126.
Lindström, Jan, Maschler, Yael, Pekarek Doehler, Simona, 2016. Grammar and negative epistemics in talk-in-interaction: cross-linguistic studies.J. Pragmat. 106, 72--79.
Linell, Per, Hofvendal, Johan, Lindholm, Camilla, 2003. Multi-unit question turns in institutional interactions: sequential organizations andcommunicative functions. TEXT 23 (4), 539--571. http://dx.doi.org/10.1515/text.2003.021
Maynard, Douglas, Heritage, John, 2005. Conversation analysis, doctor--patient interaction and medical communication. Med. Educ. 39 (4), 428--435. http://dx.doi.org/10.1111/j.1365-2929.2005.02111.x
Melander Marttala, Ulla, 1995. Innehåll och perspektiv i samtal mellan läkare och patient. En språklig ocn samtalsanalytisk undersökning (diss.).Institutionen för nordiska språk, Uppsala.
Norrby, Catrin, Wide, Camilla, Lindström, Jan, Nilsson, Jenny, 2012. Finland Swedish as a non-dominant variety of Swedish-extending the scopeto pragmatic and interactional aspects. In: Muhr, R. (Ed.), Non-dominant Varieties of Pluricentric Languages: Getting the Picture. Peter Lang,Framkfurt am Main, pp. 47--60.
Ӧstman, Jan-Ola, 1981. You Know: A Discourse Functional Approach. John Benjamins, Amsterdam.Pomerantz, Anita, 1980. Telling my side: ‘‘limited access’’ as a fishing device. Sociol. Inq. 50 (3--4), 186--198. http://dx.doi.org/10.1111/j.1475-
682X. 1980.tb00020.xRaymond, Geoffrey, 2003. Grammar and social organization: yes/no type interrogatives and the structure of responding. Am. Sociol. Rev. 68,
939--967. http://dx.doi.org/10.2307/1519752Sacks, Harvey, 1987. On the preferences for agreement and contiguity in sequences in conversation. In: Button, G., Lee, J.R.E. (Eds.), Talk and
Social Organisation. Multilingual Matters, Clevedon, pp. 54--69.Sacks, Harvey, Schegloff, Emanuel, Jefferson, Gail, 1974. A simplest systematics for the organization of turn-taking in interaction. Language 50
(4), 696--735. http://dx.doi.org/10.2307/412243Sacks, Harvey, Jefferson, Gail, Schegloff, Emanuel, 1977. The preference for self-correction in the organization of repair in conversation.
Language 53 (2), 361--382. http://dx.doi.org/10.2307/413107Schegloff, Emanuel, 1996. Turn organization: one intersection of grammar and interaction. In: Ochs, E., Schegloff, E., Thompson, S.A. (Eds.),
Interaction and Grammar. Cambridge University Press, Cambridge, pp. 52--133. http://dx.doi.org/10.1017/CBO9780511620874.002Stivers, Tanya, 2011. Morality and question design: ‘‘of course’’ as contesting a presupposition of askability. In: Stivers, T., Mondada, L., Steensig,
J. (Eds.), The Morality of Knowledge in Conversation. Cambridge University Press, Cambridge, pp. 82--106. http://dx.doi.org/10.1017/CBO9780511921674.005
Stivers, Tanya, Hayashi, Makoto, 2010. Transformative answers: one way to resist a question’s constraints. Lang. Soc. 39 (1--25.). http://dx.doi.org/10.1017/S0047404509990637
Stivers, Tanya, Robinson, Jeffrey D., 2006. A preference for progressivity in interaction. Lang. Soc. 35 (3), 367--392. http://dx.doi.org/10.1017/S0047404506060179
J. Lindström, S. Karlsson / Journal of Pragmatics 106 (2016) 129--147 147
Stivers, Tanya, Mondada, Lorenza, Steensig, Jakob, 2011. Knowledge, morality and affiliation in social interaction. In: Stivers, T., Mondada, L.,Steensig, J. (Eds.), The Morality of Knowledge in Conversation. Cambridge University Press, Cambridge, pp. 3--26. http://dx.doi.org/10.1017/CBO9780511921674.002
Tsui, Amy B.M., 1991. The pragmatic functions of ‘I don’t know’. TEXT 11 (4), 607--622. http://dx.doi.org/10.1515/text.1.1991.11.4.607Weatherall, Ann, 2011. ‘I don’t know’ as a pre-positioned hedge. Res. Lang. Soc. Interact. 44 (4), 317--337. http://dx.doi.org/10.1080/
08351813.2011.619310
Jan Lindström is a professor of Scandinavian languages at the University of Helsinki. He has done extensive research on grammar in interactionfocusing on Swedish. From a comparative point of view, he has studied the syntax of spoken Swedish and German as well as language contactbetween Finland Swedish and Finnish talk-in-interaction. He is currently a chief investigator in two large-scale projects: the Finnish Center ofExcellence in Research on Intersubjectivity in Interaction (2012--2017) and Interaction and Variation in Pluricentric Languages: CommunicativePatterns in Sweden Swedish and Finland Swedish (2013--2020).
Susanna Karlsson has a Ph.D. in Scandinavian Languages, and currently works as a language planner at the Language Council of Sweden. Herresearch mainly focuses on language planning issues for majority languages and on the relationship between interaction and grammar. Recentpublications include ‘Transfer of telephone conversations as a transition between call-takers’ (Journal of Pragmatics, 2016), ‘Subjectivity incontrast: A cross-linguistic comparison of ‘‘I think’’ in Australian English, French and Swedish’ (with K. Mullan, in Subjectivity in Language and inDiscourse; Emerald, 2012), and ‘Managing troubles-talk in the re-negotiation of a loan contract’ (with M. Ekström and A. Lindström; DiscourseStudies, 2013).